The strategy of using partial dopamine (DA) agonists as antidopaminergic antipsychotic agents is based on two sets of observations: first, that DA neurons have autoreceptors which function to decrease DA synthesis, release and neuronal firing, and hence mediate antidopaminergic signals; and, second, that partial agonists have full affinity but reduced intrinsic activity at DA receptors, consequently they exert relatively lower receptor stimulation in competition with the natural neurotransmitter, dopamine. The investigators have suggested that antipsychotic treatment with partial DA agonists, compared to treatment with DA receptor antagonists may have significant clinical advantages. Efficacy and side effects of partial DA agonists will depend on the level intrinsic activity of the partial agonist, a range which can extend from less than 10 percent to over 90 percent. The investigators work to date (see Progress Report) suggests that a partial agonist activity somewhat below 40 percent may be optimal for schizophrenia. The investigators most useful partial DA agonist is (-)-3PPP. The strategy of combining a small proporb'on of a full antagonist (e.g. haloperidol or clozopine) with the partial agonist (-)-3PPP, to produce a functionally lower intrinsic activity of (-)-3PPP, adds flexibility to the clinical testing of the partial agonist strategy. The investigators first clinical study will focus on testing the antipsychotic action of a very low dose of haloperidol (0.5 haloperidol + (-)-3PPP (flexible dose range) or (-)3PPP placebo, compared to a third active control arm (5 mg bid haloperidol + (-)-3PPP placebo). Treatment in each of these three aims will be evaluated in the three primary symptom cluster of schizophrenia: 1) hallucinations/delusions; 2) disorganization, and 3) negative symptoms, and on cognitive function. Our second study will be designed exactly like the first study, except that a very low dose of clazopine will be utilized. Clazapine is low affinity tigand and may, in combination with (-)-3PPP, produce a better antipsychotic action based on its greater displacability. Because the investigators have already demonstrated significant efflcacy, but efflcacy to which tolerance occurs, the investigators current goal is to demonstrate that the efficacy of (-)-3PPP treatment, by modifying intrinsic activity and dosing schedules can be extended. If the investigators are able to overcome the efficacy tolerance, then one of these treatments would be ready for broader multicenter testing. Application of this strategy to other syndromes where neuroleptics improve psychosis is also indicated. In the biochemistry laboratory, we propose to use the D2', D4 2. D4 4 and D4 7 cloned receptors in cultured CHO cells to examine the intrinsic activity and the desensitization potential of different DA agonists at the human D2-family receptors, in vitro. Cloned receptor desensitization, change in GTPyS activation, and arachidonic acid release will be used to study the mechanisms of tolerance and desensitization operating here. These data will predict and later help us to select optimal agonist intrinsic activity ior clinical testing for antipsychotic actvity, duration of therapeutic action, and side effects.